https://github.com/reality361/opencv-object-detection-ui
🚀 Object Detection with OpenCV & YOLO | Interactive UI 🎯
https://github.com/reality361/opencv-object-detection-ui
object-detection opencv python yolo yolov5
Last synced: about 1 month ago
JSON representation
🚀 Object Detection with OpenCV & YOLO | Interactive UI 🎯
- Host: GitHub
- URL: https://github.com/reality361/opencv-object-detection-ui
- Owner: Reality361
- License: mit
- Created: 2025-03-13T19:01:02.000Z (over 1 year ago)
- Default Branch: main
- Last Pushed: 2025-03-13T19:14:25.000Z (over 1 year ago)
- Last Synced: 2025-03-13T20:26:37.742Z (over 1 year ago)
- Topics: object-detection, opencv, python, yolo, yolov5
- Language: Python
- Homepage:
- Size: 21.1 MB
- Stars: 1
- Watchers: 1
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
- License: LICENSE
Awesome Lists containing this project
README
# 🚀 Object Detection with OpenCV & YOLO | Interactive UI 🎯
🔍 **Real-time Object Detection using YOLO and OpenCV** with a **scrollable, interactive UI** for selecting object categories dynamically.
---
## 📸 **Demo**
---
## 🎯 **Project Overview**
This project implements **real-time object detection** using **YOLO (You Only Look Once)** and **OpenCV** in Python. It features:
✅ **Deep Learning-based Object Detection (YOLOv4-tiny)**
✅ **Customizable Object Selection via UI Buttons**
✅ **Scrollable Category Panel** (Supports 80+ object categories)
✅ **Optimized UI** (Compact, modern, and business-ready)
✅ **Fast and Efficient** (Utilizing OpenCV’s DNN module)
---
## 📂 **Project Structure**
```
opencv-object-detection-ui/
├── dnn_model/ # YOLO model files
│ ├── yolov4-tiny.cfg
│ ├── yolov4-tiny.weights
│ ├── classes.txt
├── main4.py # Main object detection script
├── gui_buttons.py # GUI button system with scrolling support
├── README.md # Project documentation
```
---
## 🛠️ **Installation & Setup**
### 🔹 1. Clone the Repository
```bash
git clone https://github.com/Reality361/opencv-object-detection-ui.git
cd opencv-object-detection-ui
```
### 🔹 2. Install Dependencies
Ensure you have installed dependencies about OpenCV.
📌 **Dependencies include:**
- `opencv-python`
- `numpy`
### 🔹 3. Download YOLO Weights
The project requires **YOLOv4-tiny pre-trained weights**.
📥 Download them from: [YOLO Weights](https://github.com/AlexeyAB/darknet/releases)
Place them inside the `dnn_model/` directory:
```
dnn_model/
├── yolov4-tiny.cfg
├── yolov4-tiny.weights
├── classes.txt
```
---
## ▶️ **Running the Object Detector**
```bash
python main4.py
```
📌 **Controls:**
- **Click on buttons** to enable/disable object detection for specific categories.
- **Scroll with the mouse wheel** to navigate the object selection panel.
- **Press `ESC`** to exit.
---
## 📜 **How It Works**
1. **Loads the YOLOv4-tiny model** via OpenCV’s `cv2.dnn_DetectionModel()`.
2. **Reads class labels** from `classes.txt`.
3. **Captures frames from the webcam** and processes them for detection.
4. **Detects objects** based on confidence threshold.
5. **Draws bounding boxes** and labels detected objects.
6. **Users can dynamically select objects** to detect via an interactive UI.
---
## 🎨 **UI Features**
- **Scrollable Button Panel** → All object categories fit neatly.
- **Compact Button Layout** → Easy readability & interaction.
- **Smooth Click & Scroll Support** → Optimized for usability.
### **Example Object Categories:**
| Object | Button Color |
|---------|-------------|
| Person | 🟦 Blue |
| Car | 🟥 Red |
| Dog | 🟩 Green |
| Bottle | 🟪 Purple |
---
## 🏗️ **How to Add More Object Categories?**
1. **Edit `dnn_model/classes.txt`**
- Add the new object name on a new line.
2. **Modify `initialize_buttons()` in `main4.py`**
- New buttons are automatically added using the categories list.
---
## 🚀 **Performance Optimization**
- Convert the model to **ONNX** for OpenCV acceleration:
```bash
python export.py --weights best.pt --include onnx
```
- Run with **GPU acceleration** (if supported):
```python
net.setPreferableBackend(cv2.dnn.DNN_BACKEND_CUDA)
net.setPreferableTarget(cv2.dnn.DNN_TARGET_CUDA)
```
---
## 📝 **TODOs & Future Improvements**
✔ Enhance UI further with dynamic styling
✔ Add support for **YOLOv5**
✔ Train a custom model for additional objects
✔ Optimize for **edge devices** like Raspberry Pi
---
## 📜 **License**
🔓 This project is open-source and available under the **MIT License**.
---
## 📬 **Contact & Support**
👨💻 **Maintainer**: [Reality361](https://github.com/Reality361)
💡 Have questions? Open an [issue](https://github.com/Reality361/opencv-object-detection-ui/issues).
🚀 Enjoyed the project? Give it a ⭐ on GitHub!